Donkey Milk Study: Milk miRNA Across Species (Zoziuk 2026)

From 29 publicly available sequencing datasets, including 4 on donkey milk, 1,549 conserved miRNAs were identified. miR-148a-3p accounted for the largest proportion in all four types of milk, averaging approximately 21.5%.

The study in brief

• Original title: Conserved Core and Species-Specific Signatures in the Milk Exosomal microRNA Targetome: A Preliminary Comparative In Silico Analysis of Human, Cow, Goat and Donkey Milk
• Authors: Zoziuk M., Djibagaou A.D., Terrinoni A. et al.
• Journal: International Journal of Molecular Sciences, 27(15), 6952 (2026)
• Country: Italy · Study type: Laboratory study

How it was investigated

Pure computer analysis without own samples: Publicly available sequencing data of small RNAs from 29 milk or milk cell samples from humans (n = 8), cows (n = 13), goats (n = 4), and donkeys (n = 4) were re-evaluated uniformly. The four donkey samples come from two published comparative studies on milk vesicles of agricultural livestock. The miRNAs were quantified against human reference sequences, so that only miRNAs conserved between the species were identified.

Important figures

• 29 – public datasets (human 8, cow 13, goat 4, donkey 4)
• 1549 – conserved miRNAs across all samples
• 456 – of which detected in all four types of milk
• ≈ 21.5% – average proportion of miR-148a-3p of the total miRNA signal
• ≈ 73.7% – proportion of the 20 most common miRNAs
• 4577 – genes after Rausch filter in the evaluation (6039 excluded)

Results

Across all 29 datasets, 1549 conserved miRNAs were found, 456 of which in all four types of milk. The spectrum was highly concentrated on a few miRNAs: miR-148a-3p was the most common miRNA in every type of milk (averaging approximately 21.5%), the five most common made up about 42.5%, the 20 most common about 73.7%. The composition of the 15 most common miRNAs was similar between the species and differed mainly in quantity.

After Rausch filtering, 4577 predicted target genes remained. 1809 of them (39.5%) formed a common core with similarly high scores in all four types of milk; with 770 genes (16.8%), the donkey-related category was the second largest, followed by the human-related one with 430 genes.

In the permutation test, the size of the donkey category with 770 genes was significantly above the expected value of chance of 325 ± 16 (p < 0.001). The scores of cow and goat correlated most strongly (ρ = 0.93), donkey and human most weakly (ρ = 0.67).

Limitations of the study

• Only computer predictions: The study does not measure an effect. It assumes that miRNAs from milk are taken up and regulate genes in human cells – an assumption that, according to the authors themselves, is scientifically controversial. Statements about effects in humans cannot be derived from this.
• Only four donkey samples: The donkey data are based on four datasets from two earlier projects, one of which had a very low assignment rate (6.68%). Breed, lactation stage, and preparation differ between the source datasets and are only listed in the appendix.
• Only conserved miRNAs: Since all read sequences were evaluated against human references, donkey- or ruminant-specific miRNAs are missing. Precisely the species-specific differences are thus underestimated according to the authors.
• Mixed starting samples: The datasets include mature milk, colostrum (ten of the 13 cow samples), and milk fractions or milk cells, isolated with different methods. The data availability additionally lists three GEO series that are not assigned in the sample description of the methods section.
• Self-set thresholds: Categories are created over freely chosen boundaries (1.5-fold distance, 60th percentile). Average scores and fold-changes in Table 1 are based only on the 50 strongest genes per category, the gene numbers on all 4577 genes.

Level of evidence: Basic research (laboratory) – describes properties of milk, no statement about effects in humans.

Our assessment

The work is a computer analysis of already published data and contains no own milk samples. It describes which conserved miRNAs were common in the evaluated datasets and how calculated target gene profiles differ between milk types. Statements about effects on genes, cells, or the health of humans cannot be derived from this.

Source: Zoziuk M., Djibagaou A.D., Terrinoni A., Koroliouk D., Colizzi V. (2026). Conserved Core and Species-Specific Signatures in the Milk Exosomal microRNA Targetome: A Preliminary Comparative In Silico Analysis of Human, Cow, Goat and Donkey Milk. International Journal of Molecular Sciences. DOI: 10.3390/ijms27156952 · Summarized by the DON DONKEY'S editorial team, as of October 2026.

End of the study summary. The study was conducted independently of DON DONKEY'S® and does not refer to our products. Our donkey milk is a food.